While rheumatoid arthritis is primarily considered a condition affecting the joints and impairing function, past data suggest that cardiac disease represents the number one cause of mortality in rheumatoid arthritis. However, the adjusted rates of cardiovascular death and myocardial infarction in rheumatoid arthritis are poorly characterized. Additionally, whether the increased cardiovascular risk is because of the medications used for rheumatoid arthritis or the underlying disease severity is unknown. The proposed research has two major aims: 1) to quantify the rates of cardiovascular death and myocardial infarction in patients with rheumatoid arthritis after controlling for known cardiovascular risk factors and 2) to determine the contribution of rheumatoid arthritis medication exposure and disease severity to cardiovascular disease rates. Prior work on this issue has largely been conducted in referral populations and attempts to control for known cardiovascular risk factors have been poor. We propose to study this issue in a large Medicare/Medicaid database that we have extensive experience working with. This database contains information on over 2 million patients followed for 10 years and includes diagnoses and procedures for all physician and inpatient visits. As well, prescription data from a large pharmacy benefits program has been integrated into this database allowing for a complete characterization of an individual patients medication exposure. While any one diagnosis of rheumatoid arthritis may not be accurate in such a database, the project entails a validation sub-study to develop an algorithm for selecting patients with a high probability of having rheumatoid arthritis. The proposed project will be an important advance in this area because of the large number of patients with rheumatoid arthritis to be included (over 5,000), the community-based nature of their care, the ability to control for known cardiovascular risk factors, the extensive medication information allowing for us to explore key hypotheses regarding corticosteroid exposure, and the attempt to simultaneously control for disease severity